Risk of Lower Extremity Amputations in People with Type 2 Diabetes Mellitus Treated with Sodium-glucose-co-transporter-2 Inhibitors in the USA: A Retrospective Cohort Study

From Wiki Journal Club
Jump to: navigation, search
Yuan Z, et al. "Risk of lower extremity amputations in people with type 2 diabetes mellitus treated with sodium-glucose co-transporter-2 inhibitors in the USA: A retrospective cohort study". Diabetes Obesity & Metabolism. 2018. 20(3):582-589.
PubMedFull text

Clinical Question

In patients with type 2 diabetes mellitus (T2DM), does treatment with sodium glucose co-transporter 2 (SGLT2) inhibitors overall, and canagliflozin specifically, increase the incidence of amputation compared with non-SGLT2 inhibitor antihyperglycaemic agents (AHAs)?


This retrospective cohort study of patients with T2DM found no evidence suggesting new users of canagliflozin have an increased risk of below-knee lower extremity (BKLE) amputations compared to new users of non-SGLT2 inhibitor AHAs.

Major Points

Two trials called the CANagliflozin cardioVascular Assessment Study (CANVAS) and CANVAS-Renal (CANVAS-R) studied the effects of canagliflozin in patients with T2DM and a history or elevated risk of cardiovascular disease. The data was compiled and analyzed as part of the CANVAS Program, where results from a secondary outcome analysis suggested a greater risk of amputation in patients treated with canagliflozin.

This research report is a further evaluation of the general population of patients with type 2 diabetes. The results of this real-world study showed no increased risk of BKLE amputations in T2DM patients treated with canagliflozin as compared to those treated with non-SGLT2 inhibitor AHAs.

There are several dissimilarities between the CANVAS Program and this observational study, possibly leading to the difference in outcomes. The mean age of participants was approximately 10 years lower in the retrospective study than in the CANVAS Program. Due to the observational nature of this study, many risk factors for amputation and diabetes control such as duration of diabetes and BMI could not be identified and compared to those reported by the CANVAS Program.


According to the American Association of Clinical Endocrinologists February 2017 guidelines, SGLT2 inhibitors are an acceptable alternative to metformin for initial T2DM therapy. In May 2017, a safety announcement was issued by the FDA warning of an increased risk of leg and foot amputations based on the findings from the CANVAS Program. This resulted in the addition of a Boxed Warning to canagliflozin labels.

Study Design

  • Observational, retrospective, new-user cohort study
  • N= 346,190
    • New users of SGLT2 inhibitors (n=119,567)
      • New users of canagliflozin, specifically (n=73,024)
    • New users of non-SGLT2 inhibitor AHAs (n=226,623)
  • Setting: All patients were treated in the USA.
  • Enrollment: Eligible patients were identified based on insurance claims data found through the Truven MarketScan CCAE database. Patients must have been started on an SGLT2 inhibitor or a non-SGLT2 inhibitor AHA between April 1, 2013 and October 31, 2016.
  • Mean follow-up: 1.27 years with canagliflozin and 1.04 years with non-SGLT2 inhibitor AHAs.
  • Analysis:
    • Intent-to-treat (i.e. any new user of SGLT-2 inhibitor during study period, regardless if the person remained on the drug, was considered as a SGLT-2 inhibitor user)
    • Cox proportional hazards model was used for comparative analysis
  • Primary endpoint: The incident case of BKLE amputation, as defined by a procedural code in the Truven MarketScan CCAE database.


Inclusion Criteria

  • ≥ 365 days of observation prior to the cohort start date (i.e. index date)
  • Diagnosis of T2DM
  • Newly exposed to a SGLT2 inhibitor (canagliflozin, dapagliflozin, or empagliflozin) or a non-SGLT2 inhibitor AHA (non-metformin) between April 1, 2013 and October 31, 2016.

Exclusion Criteria

  • Patients with a history of type 1 diabetes
  • Patients with secondary diabetes prior to or on the date of exposure
  • Previous exposure to any SGLT2 inhibitors during the observation period

Baseline Characteristics

  • All baseline characteristics were well balanced after EPS matching
  • Mean age (years):
    • Canagliflozin: 53.2
    • Non-SGLT2 inhibitor AHA: 53.3
  • Sex (%):
    • Canagliflozin
      • Male: 55.5
      • Female: 44.5
    • Non-SGLT2 inhibitor AHA
      • Male: 55.3
      • Female: 44.7
  • Medications of interest (%):
    • Agents acting on the renin-angiotensin system:
      • Canagliflozin: 74.9
      • Non-SGLT2 inhibitor AHA: 75.3
    • Calcium channel blockers:
      • Canagliflozin: 65.2
      • Non-SGLT2 inhibitor AHA: 65.8
    • Β-blocking agents:
      • Canagliflozin: 50.6
      • Non-SGLT2 inhibitor AHA: 51.1
    • HMG-CoA reductase inhibitors:
      • Canagliflozin: 79.1
      • Non-SGLT2 inhibitor AHA: 82.1
    • Diuretics:
      • Canagliflozin: 79.7
      • Non-SGLT2 inhibitor AHA: 80.1
    • Drugs for acid-related disorders:
      • Canagliflozin: 54.3
      • Non-SGLT2 inhibitor AHA: 54.6
    • Digoxin:
      • Canagliflozin: 0.5
      • Non-SGLT2 inhibitor AHA: 0.5
    • Anti-inflammatory and anti-rheumatic products, non-steroids:
      • Canagliflozin: 45.0
      • Non-SGLT2 inhibitor AHA: 45.4
    • Selective serotonin reuptake inhibitors:
      • Canagliflozin: 13.9
      • Non-SGLT2 inhibitor AHA: 14.0
  • AHA therapies (%):
    • Metformin:
      • Canagliflozin: 81.1
      • Non-SGLT2 inhibitor AHA: 76.0
    • DPP-4 inhibitors:
      • Canagliflozin: 35.9
      • Non-SGLT2 inhibitor AHA: 8.3
    • GLP-1 agonists:
      • Canagliflozin: 4.5
      • Non-SGLT2 inhibitor AHA: 0.3
    • Thiazolidinediones:
      • Canagliflozin: 9.8
      • Non-SGLT2 inhibitor AHA: 3.7
    • Sulfonylureas:
      • Canagliflozin: 40.5
      • Non-SGLT2 inhibitor AHA:15.2
    • Insulins and analogues:
      • Canagliflozin: 6.6
      • Non-SGLT2 inhibitor AHA: 2.7
    • Other AHAs:
      • Canagliflozin: 2.2
      • Non-SGLT2 inhibitor AHA: 0.6


Primary outcome

  • The incident case of BKLE amputation, as defined by a procedural code in the Truven MarketScan CCAE database.
  • Incidence rate of BKLE amputation event post exposure (per 1000 person-years):
    • Canagliflozin: 1.18
    • Non-SGLT2 inhibitor AHAs: 1.12
    • Hazard ratio (HR) 0.98 [95% CI 0.68, 1.41]; P=0.92, calibrated P=0.95


  • The populations studied in the CANVAS Program were not similar to those analyzed in this observational research. Differences between groups that may influence risk of amputation and poorer diabetes care include:
    • Socio-economic status
    • Behavioral variables
    • Body weight
    • HbA1c
    • Duration of diabetes
    • Age less than 65 (with commercially-available insurance through an employer)
  • Insurance databases such as the Truven MarketScan CCAE database were created for financial purposes instead of for research.
  • The lower rate of amputation in this study compared to the CANVAS Program may have been due to differences in baseline characteristics. For example, patients in this study had a mean age of 53.2 in the canagliflozin group and 53.3 in the non-SGLT2 inhibitor AHA group. The mean age of participants in the Canvas Program was 63.3 years.
  • Baseline use of AHAs remained imbalanced after EPS matching.


This study was funded by Janssen Research & Development, LLC.

Further Reading

Neal B, Perkovic V, Mahaffey KW, et al. Canagliflozin and Cardiovascular and Renal Events in Type 2 Diabetes. N Engl J Med. 2017. doi:10.1056/NEJMoa1611925.